The trusted source for
healthcare information and
By Robert J. Nardino, MD, FACP
To touch or not to touch—that is the question facing parents and caregivers of premature and low birth weight infants. These infants are exposed to a constant barrage of light and sound while they are in Neonatal Intensive Care Units (NICU), in what must be a shocking change from the nurturing environment of the womb. Some have postulated that massage, by reducing the stress caused by the NICU, can promote infant growth and development.
Touch, stretch, and motion stimuli for a nine-month term are thought to be important for infant development, and the premature infant loses access to them abruptly. The lack of usual tactile, visual, and auditory stimuli can result in an impaired intellectual level.1
Mechanism of Action
The mechanism by which infant massage may provide benefit is not understood clearly. Reductions in cortisol levels with massage have been reported, but it is unclear whether this plays an important role.2 Others have not seen this reduction in cortisol, but have demonstrated an enhancement in the activity of the sympathetic nervous system.3 Improved cognitive development with massage may arise from increased catecholamine activity, resulting in heightened attention and improved memory. Increased secretion of growth hormone also has been reported, suggesting a third hypothesis.4
Massage does not appear to result in increased caloric intake.5 However, there has been some suggestion that weight gain might result from better conversion of food into energy for growth.
There are some variations in the intervention, but the most commonly used regimen is outlined in Table 1. Infant massage entails a systematic stroking, gently but with pressure.
Sample massage intervention
|Three 15-minute periods during three consecutive hours, daily for
|Components of the 15-minute session:|
|1. Five-minute tactile phase: infant placed in prone position,
stroked with flat of fingers of both hands (using a gentle pres-
sure, but not too light as to create a tickling stimulus), one
minute over each region in sequence:
| • 12 five-second strokes from top of head, down side of face
to neck, back to head
| • 12 five-second strokes from back of neck across shoulders
and back to neck
| • 12 five-second strokes from the upper back down to the
waist and back up
| • 12 five-second strokes from the thighs down to the ankles
and back to thighs
| • 12 five-second strokes from the shoulders to the wrists,
back to shoulders
|2. Five-minute kinesthetic phase: infant placed in supine posi-
tion, six passive flexion-extension movements lasting 10 sec-
onds to right arm, left arm, right leg, left leg, both legs
|3. Five-minute tactile phase identical to phase one|
|Adapted from: Field TM, et al. Tactile/kinesthetic stimulation
effects on preterm neonates. Pediatrics 1986;77:654-658.
In 1969, Solkoff published the first study of massage therapy in 10 preterm infants. The intervention was just five minutes, shorter in duration than the sample massage intervention described in Table 1. Very few data are provided.6 Several studies at various locales followed, and form the basis for the reviews described below.
A systematic review by Ottenbacher et al in 1987 concluded that interpretation of the studies of tactile stimulation could not be accurately undertaken because the studies were heterogeneous in design, outcome measure, and quality.7 The authors’ analysis included all studies up to that time, including uncontrolled studies. The effect size was much larger in studies that were not randomized properly. Ireland and Olson recently reviewed the existing literature on massage and therapeutic touch in children.8 They looked at seven studies of massage therapy in preterm infants who did not have other underlying conditions and found consistent beneficial effects on weight gain, activity level, and length of hospitalization. However, while systematic in their search for all relevant studies, they did not attempt a quantitative analysis of the data.
The most thorough and quantitative approach to the body of literature of massage therapy can be found in the Cochrane Collaboration Review published in March 2000.9 All relevant studies were identified and assessed for the strength of their methods. Thirteen studies were ultimately included for analysis, and their outcomes were combined using weighted averages. Because different studies measured different outcomes, some of the analyses are based on small numbers of patients.
Table 2 shows a summary of differences in selected outcomes between infants receiving massage and infants in control groups. There are statistically significant improvements in several outcomes, including daily weight gain, length of stay, and measures of development. Massage, performed one to three times daily for between four and 10 days, resulted in weight gain ranging from 3-8 g/d in excess of what the control group gained. Benefits in motor and behavioral development were much less consistent. However, the overall interpretation was that there is insufficient evidence to support the use of massage, primarily because the methodologic quality of the studies is weak.
Selected outcomes from studies comparing massage vs. routine care in preterm/low birth weight infants
|Outcome||# of Studies||# of Patients (Exp/Control)||Weighted Mean Difference* (95% CI)||Clinically Meaningful||Consistency|
|Daily weight gain5,12,13,15-17||6||274 (139/135)||5.1 (3.5 to 6.7)||yes||5/6 favored massage|
|Length of stay5,12,13,16-18 (days)||7||206 (103/103)||-4.6 (-6.6 to -2.6)||yes||3/7 favored massage|
|Brazelton Scale: habituation5,12,13||3||103 (51/52)||0.8 (0.5 to 1.1)||no||2/3 favored massage|
|Brazelton Scale: motor maturity5,12,13||3||110 (55/55)||0.8 (0.5 to 1.1)||perhaps 1/3 favored massage|
|Weight at 4-8 mo. follow-up5,10 (oz)||2||49 (25/24)||0.5 (-0.05 to 1.1)||no||1/2 favored massage|
|*Studies were weighted based on strength of design|
|Adapted from: Vickers A, et al. Massage for promoting growth and development of preterm and/or low birth-weight infants. Cochrane Database Syst Rev 2000;2:CD000390.|
Scafidi and Field randomized 28 infants born to HIV-positive mothers to massage or control.11 These infants were not premature or low birth weight. They found similar results to those in preterm infants—improved average daily weight gain and improvements on Brazelton Scale measurements. A study that investigated the effects of massage on cocaine-exposed preemies was included in the systematic reviews.12
Scafidi and colleagues also attempted to determine if certain subsets of infants fared better with massage than others.13 The idea was to direct the intervention to those patients who most likely would benefit. They found that infants who had higher scores on the Obstetric Complications Scale derived more benefit from massage.
Many of the published studies evaluating massage are of low quality. Even among studies that are randomized, there is no blinding to treatment allocation. Also, the control intervention or usual care is not well defined, and there is no description of efforts to ensure that patients were treated the same in all respects other than the experimental intervention. There is no standardization as to who is administering the intervention. Length of stay data were not reported in several of the studies; failure to report these data suggests that there was no difference, meaning that the results from the studies that did report them may be overestimating the effect.
No adverse effects were noted in the published studies. However, there have been observations that marginally stable infants occasionally collapse and develop respiratory distress following a massage session.14 It is unclear whether there is a threshold for overstimulation and whether there are any significant lasting consequences.
Overall, it is difficult to draw conclusions from the literature on massage in premature and low birth weight infants. The data suggest a possible improvement in short-term weight gain. Data about length of stay, while of importance, may have been compromised by selective reporting. Confirmatory studies with strict methodology to reduce bias and long-term follow-up are needed before massage can be widely adopted. At the same time, it is a noninvasive intervention that in general appears safe to implement. Parents or other non-skilled personnel such as volunteers can administer this intervention. It would seem obvious that premature infants should experience touch, although it is interesting that in the studies that made the comparison, massage yielded benefits in growth that gentle, still touch did not.
The data favoring the use of massage in preterm infants, though weak, do indicate an increase in daily weight gain over the short periods it has been studied. Additionally, massage for any purpose is unlikely to be harmful. Therefore, massage should be considered as an adjunct in the management of premature infants, with the expectation of modest increases in daily weight gain. Whether there are benefits for other aspects of development is inconclusive. It is also uncertain whether it is cost-effective for NICU nurses to administer massage, but it can be performed readily by parents or volunteers. Infants should be medically stable before receiving massage.
Dr. Nardino is Program Director, Internal Medicine Residency, Hospital of Saint Raphael, New Haven, CT, and Assistant Clinical Professor of Medicine, Yale University School of Medicine.
1. Scarr-Salapatek S, Williams ML. The effects of early stimulation on low birth weight infants. Child Dev 1973;44:94-101.
2. Acolet D, et al. Changes in plasma cortisol and catecholamine concentrations in response to massage in preterm infants. Arch Dis Child 1993;68(1 Spec No):29-31.
3. Kuhn CM, et al. Tactile-kinesthetic stimulation effects on sympathetic and adrenocortical function in preterm infants. J Pediatr 1991;119:434-440.
4. Schanberg SM, Field TM. Sensory deprivation stress and supplemental stimulation in the rat pup and preterm human neonate. Child Dev 1987;58:
5. Field TM, et al. Tactile/kinesthetic stimulation effects on preterm neonates. Pediatrics 1986;77:654-658.
6. Solkoff N, et al. Effects of handling on the subsequent development of premature infants. Dev Psychol 1969;1:765-768.
7. Ottenbacher KJ, et al. The effectiveness of tactile stimulation as a form of early intervention: A quantitative evaluation. J Dev Behav Pediatr 1987;8:68-76.
8. Ireland M, Olson M. Massage therapy and therapeutic touch in children: State of the science. Altern Ther Health Med 2000;6:54-63.
9. Vickers A, et al. Massage for promoting growth and development of preterm and/or low birth weight infants. Cochrane Database Syst Rev 2000;2:
10. Rice RD. The effects of the Rice infant sensorimotor stimulation treatment on the development of high-risk infants. Birth Defects Orig Artic Ser 1979;15:7-26.
11. Scafidi F, Field T. Massage therapy improves behavior in neonates born to HIV-positive mothers. J Pediatr Psychol 1996;21:889-897.
12. Wheeden A, et al. Massage effects on cocaine-exposed preterm neonates. J Dev Behav Pediatr 1993;14:
13. Scafidi FA, et al. Factors that predict which preterm infants benefit most from massage therapy. J Dev Behav Pediatr 1993;14:176-180.
14. Feldman R, Eidelman AI. Intervention programs for premature infants. How and do they affect development? Clin Perinatol 1998;25:613-626, ix.
15. Adamson-Macedo E. Effects of tactile stimulation on low and very low birthweight infants during the first week of life. Curr Psychol Res Rev 1985;4:305-308.
16. White JL, Labarba RC. The effects of tactile and kinesthetic stimulation on neonatal development in the premature infant. Dev Psychobiol 1976;9:569-577.
17. White-Traut RC, Tubeszewski KA. Multimodal stimulation of the premature infant. J Pediatr Nurs 1986;1:90-95.
18. White-Traut RC, et al. Developmental intervention for preterm infants diagnosed with periventricular leukomalacia. Res Nurs Health 1999;22:131-143.